VMware ESXi and vSphere Cluster Management

How to Revert a Virtual Machine to a Snapshot in VMware vSphere

Learn how to safely revert a VMware vSphere virtual machine to a snapshot, understand data loss, validate the rollback, and troubleshoot snapshot operations.

Reverting a virtual machine (VM) to a snapshot restores it to the state represented by a selected point-in-time capture. This can undo an unsuccessful application update, test configuration, software installation, or other work performed after the snapshot was created.

Important: Reverting is a potentially destructive rollback. Disk changes made after the selected snapshot are removed from the VM's active state. Confirm the VM, snapshot name, description, timestamp, and expected data-loss impact before approving the operation.

Key VMware Terms

Virtual machine (VM): A software-defined computer managed by vSphere.
VMware ESXi: The hypervisor that runs VMs on a physical host.
vCenter Server: The centralized VMware management platform commonly used to administer ESXi hosts, VMs, datastores, and snapshots.
vSphere Client: The browser-based administrative interface used to manage a VMware environment through vCenter Server or, in some cases, an ESXi host.
Snapshot: A point-in-time capture of VM state intended primarily for short-term rollback.
Snapshot tree: The visual hierarchy of snapshots associated with a VM.
Revert: An operation that restores a VM to the state represented by a chosen snapshot.
Delta disk: A file that stores disk writes made after a snapshot is created.
Memory snapshot: Snapshot information that records the VM's memory state so it can potentially return to the captured execution point.
Quiescing: Coordinating guest filesystem or application activity to improve snapshot consistency.
Consolidation: Merging remaining snapshot delta files when VMware reports that snapshot disks need consolidation.

What Reverting to a Snapshot Does

When a snapshot is created, VMware records a recovery point for the VM. Subsequent disk writes are directed through snapshot-related delta disks. A revert changes the active VM state back to the selected snapshot and discards the later disk changes from that active state.

Changes that may be lost include files created or modified after the snapshot, application installations or updates, operating system changes, registry or service changes, and other configuration work. A revert is therefore broader than undoing one file or one setting.

If the snapshot included memory, VMware also captured the VM's running memory state. Depending on the operation and environment, the VM may return to the running state represented by that capture rather than simply booting from the restored virtual disks. A snapshot without memory represents disk and VM configuration state, but not the contents of the guest's active memory.

Revert, Create, Delete, and Consolidate Compared

Action	Effect on VM state	Effect on post-snapshot disk changes	Typical use case	Primary risk or consideration
Create snapshot	Creates a point-in-time rollback reference.	Later writes are tracked in delta disks.	Protecting a VM before a planned change or creating a lab baseline.	Snapshots consume storage and should not replace backups.
Revert to snapshot	Restores the VM to the selected snapshot state.	Later active-state changes are discarded.	Undoing a failed update, test, or configuration change.	Recent files, applications, and configuration changes may be lost.
Delete snapshot	Normally preserves the current VM state.	Snapshot delta data is merged or committed into the appropriate disk chain.	Removing a temporary rollback point after validation.	Deletion can require time, datastore space, and I/O; it does not normally revert the VM.
Consolidate snapshots	Merges remaining snapshot-related delta files into the expected disk structure.	Residual delta files are committed without intentionally rolling the VM back.	Resolving a snapshot consolidation warning.	Check datastore free space and perform the operation under suitable change control.

Prerequisites and Pre-Revert Preparation

Before starting, identify the VM and the exact recovery point. VMware administrators should understand basic ESXi administration, VM power operations, snapshot concepts, backup strategy, and role-based permissions.

Identify the target VM. Confirm its inventory name, folder, host, datastore, and—when relevant—business owner. Similar VM names can make a wrong selection easy.
Confirm permissions. Verify that your account has permission to manage snapshots for the VM. If the option is unavailable, review the assigned vCenter role and scope using your organization's access-control process.
Review the snapshot tree. Examine the snapshot name, description, timestamp, and position in the snapshot hierarchy. Similar timestamps or multiple branches require extra care.
Protect recent data. Determine whether files, database transactions, logs, or other data created after the snapshot need an independent backup or export before rollback.
Coordinate the change. Use a maintenance window for business-critical systems. Notify users and application owners, and stop or protect workload activity as appropriate.
Assess consistency. A crash-consistent snapshot may not represent a clean application state. For important workloads, consider whether guest quiescing or application-consistent protection was used and whether the application can safely be rolled back.
Check infrastructure health. Confirm that the VM, ESXi host, vCenter Server, and datastore are accessible and that sufficient datastore capacity is available.
Check for conflicting operations. Ensure that no snapshot creation, deletion, consolidation, migration, backup, or other VM task is already in progress.
Prepare validation. Decide which power-state, operating-system, application, file, connectivity, and service checks will prove that the rollback succeeded.

Pre-Revert Checklist

Check	Why it matters	Completed
Correct VM identified	Prevents rolling back a similarly named or unrelated VM.	☐
Correct snapshot identified	Determines exactly which VM state will be restored.	☐
Snapshot timestamp and description reviewed	Helps distinguish similar recovery points.	☐
Current data protection considered	Provides a recovery path for data that may be discarded.	☐
Users or application owners notified	Prevents unexpected interruption and coordinates validation.	☐
No conflicting snapshot task is running	Reduces task conflicts and incomplete operations.	☐
Post-revert validation plan prepared	Ensures the VM is not returned to users before it is tested.	☐

How to Revert a VM in the vSphere Client

The exact labels vary slightly between vSphere Client releases and inventory views, but the workflow is consistent. The operation is normally performed through vCenter Server, which centrally manages the VM and its ESXi host.

Sign in to the vSphere Client. Use an account authorized to manage snapshots.
Locate the VM. Open the inventory and select the target VM. Verify its identity and current power state.
Open snapshot management. Use the VM's snapshot view, Snapshots tab, or Snapshot Manager entry. In some client layouts, this is available from the VM's actions menu.
Inspect the snapshot tree. Expand the hierarchy if necessary. Select the intended snapshot and review its name, description, date, time, and relationship to other snapshots.
Start the revert action. Choose the control labeled Revert, Revert to, or an equivalent snapshot rollback action for the selected recovery point. Do not use Delete unless you intend to remove the snapshot while preserving the current state.
Read the confirmation warning. Confirm that changes made after the selected snapshot can be discarded. Recheck the VM and snapshot selection before accepting.
Approve the rollback. Confirm the operation only after validating its scope and data-loss consequences.
Monitor the task. Watch the VM's Tasks or Recent Tasks view until the revert completes. Do not begin another snapshot operation merely because the interface appears idle; verify the task result.

Practical Examples

Undoing an Unsuccessful Application Update

An administrator creates a snapshot named Before application upgrade - change 1234 before upgrading an application.
The upgrade completes but validation shows service failures.
The administrator reviews the snapshot tree and selects the pre-upgrade snapshot.
After confirming the warning, the administrator reverts the VM and waits for the task to finish.
The administrator verifies the previous application version, configuration, service status, and user connectivity.

The failed update and all other changes made after that recovery point are rolled back. Data created during the intervening period must come from an independent backup or export if it is still needed.

Restoring a Training Lab VM

A lab VM is snapshotted in a known-good baseline state. A learner installs test software and changes system settings. Before the next exercise, the administrator selects the documented baseline snapshot and reverts the VM. This is a useful short-term, repeatable workflow for temporary training environments, provided the snapshot does not become a substitute for backups.

Avoiding a Wrong-Snapshot Rollback

Suppose a VM has several snapshots with similar timestamps. The administrator should compare names, descriptions, timestamps, and the snapshot hierarchy before selecting one. The confirmation warning is a final safety check, not a replacement for identifying the correct recovery point.

Post-Revert Validation

Verify that the task completed successfully in Tasks or Recent Tasks, and inspect related events for warnings or errors.
Confirm the VM power state and guest operating-system state are expected.
Check that the expected application version, files, services, and configuration have returned.
Validate network connectivity, DNS or directory integration, storage access, monitoring, and other critical dependencies.
Ask the application owner or test operator to perform functional checks before returning the VM to users.
Decide whether to retain the snapshot for another short-term rollback or remove it after the environment is stable.
Record the selected snapshot, reason for rollback, operator, time, observed impact, and validation results in the change record when required.

Snapshot Management and Safety Guidance

Snapshots are short-term rollback points, not replacements for backups. They can depend on the VM's disk chain and datastore availability, while a backup strategy is designed to provide independent recovery. Review vSphere permissions and your organization's backup and change-management procedures before using snapshots on production workloads.

Use descriptive names and descriptions that include the reason, change reference, owner, and intended retention period. Remove temporary snapshots when they are no longer needed. Long-lived snapshots can grow as the VM writes data, consume datastore capacity, lengthen snapshot operations, and complicate performance and management.

Deleting a snapshot normally commits its delta data and preserves the VM's current state; it is not the same as reverting. If vSphere reports that snapshot disks need consolidation, review tasks, events, alarms, and available datastore space before performing the appropriate consolidation workflow. For broader datastore planning, see datastore considerations in vSAN.

For business-critical workloads, coordinate with application owners and use a maintenance window. Consider quiescing and application consistency rather than assuming that every snapshot is safe for databases or transactional applications. Snapshot rollback should be assessed as a whole-workload change, not treated as a precise undo function.

Troubleshooting Revert Operations

The Intended Snapshot Cannot Be Found

Possible causes include prior deletion of the snapshot, selecting a different VM, or viewing the wrong inventory object or snapshot-management screen.

Verify the VM identity and inspect its complete snapshot list and tree.
Review recent tasks and events for snapshot deletion or other management activity.
If the recovery point no longer exists, use an available backup or another approved recovery method.

The Revert Option Is Unavailable or Fails to Start

Insufficient snapshot permissions, another VM operation, or an accessibility problem involving vCenter Server, the ESXi host, the datastore, or the VM can prevent the operation.

Review the account's snapshot-management permissions and role scope.
Check Tasks, Recent Tasks, events, and alarms for a conflicting operation; wait for it to finish.
Confirm that vCenter Server, the ESXi host, the datastore, and the VM are reachable and healthy.
Do not repeatedly retry a failed operation without reviewing the task error and current VM state.

Recent Files or Application Changes Are Missing

This is normally expected when those changes occurred after the selected snapshot. It may also indicate that an older snapshot was selected than intended.

Compare the selected snapshot timestamp and description with the change timeline.
Restore needed newer data from an independent backup if one exists.
Improve future snapshot naming, descriptions, and change documentation.

Snapshot Warnings Appear After the Operation

Residual delta disks or a prior snapshot operation that did not complete cleanly can produce consolidation warnings.

Review vSphere tasks, events, alarms, and the current snapshot view.
Assess datastore free space before starting consolidation.
Perform the appropriate consolidation workflow under change control and monitor it to completion.

Exam-Relevant Notes

Revert restores the VM to a selected snapshot and can discard later disk changes.
Create snapshot establishes a rollback point; it does not restore anything.
Delete snapshot normally preserves the current VM state while committing snapshot data.
Consolidate merges remaining delta files and is different from reverting.
A snapshot is not an independent backup.
A snapshot that includes memory can capture a running execution state; a disk-only snapshot does not capture guest memory.
Always verify the snapshot tree, timestamp, description, and data-loss impact before confirming a revert.